Danone uses product classification and recognition for a marketing campaign
Knowing better your customers
Danone is one of the largest food companies in the world. To better know their customers, one regional marketing team launched a campaign, where they asked consumers to submit the name of the Danone product they purchased, along with the serial code. In return they would receive rewards and be invited to special events.
How to improve marketing with Machine Learning?
Easy! Instead of asking customers to fill in a boring web form, why not to let them submit a picture of their fridge with the products on it?
What better proof of purchase than a real picture of my products on my customer's fridge?
By implementing this solution, not only would Danone learned about the products it was selling, they would also get extra insights on where they were placed inside the fridge. The campaign was a huge success and user input multiplied by 3000%!
Machine learning to the rescue!
With a simple implementation of the existing Google Cloud Vision API, the application was able to:
- Check if the submitted picture was indeed an open fridge
- Classify and separate the different products on the picture
Even though Google API is very potent, it still cannot recognise the different Danone brands and products. For this, Danone decided to train a custom Machine Learning model in Tensor Flow. This custom model was trained specifically to recognise a large number of Danone products and thus, obtain the information Danone needed from its customers.
MLab, the Machine Learning specialists at your service!
If Machine Learning inspires you and you think you would like to implement a use case in your organisation, please get in touch. We are vendor agnostic and we will recommend and integrate the technology that adapts the best to your needs. If non available technologies satisfy your needs, we can always train a custom model tailored to your project.